You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Now that head_commit_row and head_commit_field are becoming materialized views for great speed, the internal tables rowset, rowset_row and rowset_row_field, and maybe blob as well, could (should?) have write-blocking triggers for insert, update and delete. They should never be modified directly. They should never change except in functions bundle_csv_import(), commit() and commit_delete(). These functions should temporarily disable the write triggers, make whatever changes, and then re-enable them, then refresh the materialized views.
The text was updated successfully, but these errors were encountered:
Big speedups. Working around what sure seems to be a PostgreSQL bug #285. If
we keep this approach, would be nice to add write-blocking triggers on the
internal rowset* tables, as they should never be modified directly, #286.
Now that
head_commit_row
andhead_commit_field
are becoming materialized views for great speed, the internal tablesrowset
,rowset_row
androwset_row_field
, and maybeblob
as well, could (should?) have write-blocking triggers for insert, update and delete. They should never be modified directly. They should never change except in functionsbundle_csv_import()
,commit()
andcommit_delete()
. These functions should temporarily disable the write triggers, make whatever changes, and then re-enable them, then refresh the materialized views.The text was updated successfully, but these errors were encountered: